Naa50/San-dependent N-terminal acetylation of Scc1 is potentially important for sister chromatid cohesion

The gene separation anxiety (san) encodes Naa50/San, a N-terminal acetyltransferase required for chromosome segregation during mitosis. Although highly conserved among higher eukaryotes, the mitotic function of this enzyme is still poorly understood. Naa50/San was originally proposed to be required for centromeric sister chromatid cohesion in Drosophila and human cells, yet, more recently, it was also suggested to be a negative regulator of microtubule polymerization through internal acetylation of beta Tubulin. We used genetic and biochemical approaches to clarify the function of Naa50/San during development. Our work suggests that Naa50/San is required during tissue proliferation for the correct interaction between the cohesin subunits Scc1 and Smc3. Our results also suggest a working model where Naa50/San N-terminally acetylates the nascent Scc1 polypeptide, and that this co-translational modification is subsequently required for the establishment and/or maintenance of sister chromatid cohesion.

The narrow dimension of the ribosome exit tunnel precludes large domain folding of the nascent protein, which creates a window of opportunity for co-translational modification of protein residues that would be otherwise inaccessible by protein folding 1 . Co-translational modifications occur in archaea, eubacteria, and eukaryotic cells. Among others, they include protein N-terminal acetylation (Nt-Ac) [2][3][4][5] , which involves the transfer of an acetyl group from acetyl-CoA to the protein alpha-amino group 4,5 . Nt-Ac is an ubiquitous modification, and partially or fully N-terminally acetylated proteins constitute approximately 50-70% of the proteome in budding yeast, 70-80% in Drosophila and 80-90% in human and Arabidopsis [6][7][8][9][10] . Recent investigations revealed that Nt-Ac might have a major influence on protein stability, complex formation, folding, and subcellular targeting 5,[11][12][13][14][15][16][17] , as well as being essential for development of multicellular organisms 18 .
The gene separation anxiety (san) encodes Naa50/San, the catalytic subunit of NatE, a highly conserved NAT with a classical GNAT fold that acetylates N-termini starting with Methionine followed by hydrophobic (Met-Leu-, Met-Phe-, etc.) or small polar amino acids (Met-Ser-, Met-Thr-, etc.) 21,[33][34][35] . It may act physically associated with the NatA subunits Naa10 and Naa15, or independently of this complex 19,33,[35][36][37][38][39] . Human and Drosophila cells mutant for Naa50/San show chromosome segregation defects during mitosis, including chromosome lagging and chromosomal bridges 36,39,40 . Interestingly, Drosophila Naa50/San is not required for mitosis in Naa50/San is crucial for the correct interaction between Scc1 and Smc3. The cohesin complex has a ring-shaped structure and it is composed of four subunits: Smc1, Smc3, Scc3/SA and Scc1/Rad21/Mcd1 51,52 . Coentrapment of sister chromatids occurs during DNA replication. Smc3 is internally acetylated by Eco1 acetyltransferase 49,50 and subsequent Dalmatian/Sororin recruitment 53 is essential for establishment of stable cohesion as it stabilizes the interaction between the N-terminal domain of Scc1 and the C-terminal domain of Smc3 54 . Eco1 acetylation of cohesin subunit Smc3 promotes cohesion by antagonizing the function of Wapl and/or by altering Smc3 head function 55,56 .
Deco, the Drosophila ortholog of Eco1, is also important for the establishment of sister chromatid cohesion and normal chromosome segregation during anaphase 36 . Since yeast Wapl mediates the dissociation between Scc1 and Smc3 subunits creating a cohesin's DNA exit gate 57 , and given the fact that depletion of Drosophila Deco specifically enhanced the adult wing phenotypes observed after depletion of Naa50/San ( Fig. 1D and E), we hypothesized that Naa50/San function was critical for the interaction between cohesin subunits Smc3 and Scc1. In order to test this hypothesis, we investigated if expression of a fusion construct between Scc1 and Smc3 58 could suppress the adult wing phenotypes of san RNAi-treated wing discs. Consistently, and although expression of this fusion construct by itself weakly impaired wing development ( Fig. 1F; see grey bar), its co-expression with san RNAi significantly suppressed the wing disc phenotypes observed after depletion of Naa50/San (compare black bars in Fig. 1F and wing phenotypes in Fig. 1G).
Dalmatian/Sororin (Dmt) mediates cohesion by antagonizing Wapl 53 . If Naa50/San function is specifically required for the interaction between Scc1 and Smc3, then overexpression of Dmt should suppress the san RNAi adult wing phenotypes. Consistently, whereas overexpression of Dmt significantly suppressed the adult wing phenotypes observed after depletion of Naa50/San (compare black bars in Fig. 1F and wing phenotypes in Fig. 1G), a mutant allele of dmt dominantly enhanced san RNAi phenotypes (compare black bars in Fig. 1F and wing phenotypes in Fig. 1G). Since Dalmatian/Sororin protein levels are not reduced after depletion of Naa50/San (Fig. 6B) than the observed phenotypes are not due to its destabilization. Altogether, these results suggest that Naa50/San is required for sister chromatid cohesion during Drosophila wing development, most likely by regulating (directly or indirectly) the correct interaction between the cohesin subunits Scc1 and Smc3. Naa50/San is a positive regulator of sister-chromatid cohesion in Drosophila S2 cells. In order to better understand the function of Naa50/San in sister-chromatid cohesion, we performed live-cell imaging of cells depleted for Naa50/San. Since L3 larvae neuroblasts mutant for san (zygotic mutants) were highly abnormal, with an extremely low mitotic index 36 and a highly aberrant ploidy (data not shown), tissue culture Drosophila Schneider 2 (S2) cells were used instead to study the mitotic function of Naa50/San. As expected, Naa50/San was important for centromeric sister chromatid cohesion in Drosophila S2 cells; its depletion by RNAi was associated with an abnormal association of Scc1 to chromatin (Supplementary Fig. 2A,B and D), an increase in the cells mitotic index (Fig. 2F), and an abnormal segregation of chromosomes during anaphase (Figs 2G and 3C). Interestingly, Naa50/San-depleted S2 cells also showed a significant amount of single chromatids (see arrowheads in Fig. 2B,D and E; see quantification in Fig. 2G; Movies S1-S6; see arrowheads in Movies S2-S4, and S6) that was suggestive of a general loss of sister chromatid cohesion after depletion of Naa50/San. The single chromatids showed the typical oscillatory behavior where they moved towards and away from the spindle poles Wing blade-specific RNAi using the Nubbin-Gal4 driver 43,44 and the UAS/Gal4 system 42 . (B) Adult wings of wild type Drosophila (Oregon R; OR), Drosophila expressing a control RNAi (mCherry RNAi), and Drosophila expressing RNAi for san in the larvae wing imaginal discs. (C) Scored adult wing phenotypic classes: class 1 (wild type wings); class 2 (weak wing phenotype); class 3 (san RNAi-like wing phenotype); class 4 (highly abnormal wings); class 5 (absence/ vestigial adult wings). (D) Quantification of Drosophila wing phenotypes expressing individual RNAi transgenes for mCherry, deco, scc1 or mau-2 (grey bars) or co-expressing san RNAi with mCherry RNAi, deco RNAi, scc1 RNAi, or mau-2 RNAi (black bars) in the larvae wing imaginal discs. (E) Representative adult wing phenotype of Drosophila co-expressing san RNAi with mCherry RNAi, deco RNAi, scc1 RNAi or mau-2 RNAi in the larvae wing imaginal discs. (F) Quantification of Drosophila wing phenotypes expressing a RNAi transgene for mCherry, a Smc3-Scc1 fusion construct 58 or overexpressing Dmt in the larva wing imaginal discs, and Drosophila without one copy of Dmt (dmt +/− ) (grey bars). Quantification of Drosophila wing phenotypes co-expressing san RNAi with mCherry, a Smc3-Scc1 fusion construct or Dmt in the larvae wing imaginal discs, and Drosophila without one copy of Dmt (dmt +/− ) expressing san RNAi in the larvae wing imaginal discs (black bars). (G) Representative adult wing phenotype of Drosophila co-expressing san RNAi with mCherry RNAi, a Smc3-Scc1 fusion construct or Dmt in the larvae wing imaginal discs, and Drosophila without one copy of Dmt (dmt +/− ) expressing san RNAi in the larvae wing imaginal discs. Phenotypic quantification of adult wings is mean ± S.D. of three independent experiments and is based on the classes described in (C) and Supplementary Fig. 1  as they were captured by microtubules and their association to their unique kinetochore was subsequently destabilized by Aurora B [59][60][61][62][63] . Identical results were obtained in Drosophila S2 cells with a distinct non-overlapping double-stranded RNA against san (Movie S4).
In order to investigate if, similarly to the adult wing phenotypes ( Fig. 1D and E), depletion of Deco could enhance the mitotic phenotypes of san RNAi-treated S2 cells, we analyzed the mitotic defects of S2 cells 72 hours after RNAi-treatment (lower depletion of Naa50/San when compared to 96 hours after RNAi-treatment ( Supplementary Fig. 3A)). Simultaneous co-depletion of Naa50/San and Deco (san RNAi and deco RNAi) significantly increased the number of cells with detectable single-chromatids (see arrowhead in Fig. 3D; see quantification in Fig. 3F; Movies S9 and 10; see arrowheads in Movie S10) and their mitotic index (Fig. 3E) when compared to control and single RNAi-treated cells (Fig. 3A-D; see quantification in Fig. 3F). Since an extended arrest in metaphase with bi-oriented chromosomes can potentially be associated with partial loss of cohesion 64,65 , we investigated if single chromatids could be detected immediately after metaphase arrest. Live-cell imaging of cells depleted for Naa50/San and co-depleted for Naa50/San and Deco showed detectable single chromatids immediately after metaphase arrest (see arrowhead in Movie S3 and S10). This suggested that the observed single chromatids resulted from bona fide cohesion defects, and were not simply due to the extended metaphase arrest. Altogether, these results suggest that Naa50/San is a positive regulator of sister chromatid cohesion, possibly by facilitating (directly or indirectly) the interaction between the cohesin subunits Scc1 and Smc3.

Naa50/San catalytic activity is required for chromosome segregation during mitosis.
Previously it was reported that reduction of human Naa50/San catalytic activity impaired its mitotic function 39 . Yet, since the reported mutant protein had a significant amount of residual catalytic activity and it was not expressed at endogenous levels, it was still unclear to what extent the mitotic function of Naa50/San relied on its catalytic activity. Since Naa50/San interacts with NatA 36,38 and its loss may partially impair the catalytic activity of NatA 19,35 , Naa50/San can potentially have functions independent of its catalytic activity. In order to investigate if Naa50/San catalytic activity is required for normal chromosome segregation during mitosis, we generated a transgenic genomic construct of the san gene locus, under the control of its own promoter for endogenous protein expression levels ( Supplementary Fig. 3B), and carrying two different amino acid mutations (R84A Y124F) that completely abrogated the in vitro catalytic activity of this enzyme (Fig. 5B). Consistent with the hypothesis that the catalytic activity of Naa50/San is essential for sister chromatid cohesion and normal mitosis, a wild-type genomic construct (g(san Wt )), but not the catalytically dead genomic construct of san (g(san R84A Y124F )), fully rescued the lethality of zygotic loss-of-function mutant alleles of san ( Fig. 4F; Supplementary Table 3), the sterility of females whose germ-line was mutant for san (germ-line clones) (Fig. 4E), and the mitotic defects of syncytial blastoderm embryos mutant for san (maternal mutants) ( Fig. 4A-D') 36,40 . These results fully support the hypothesis that Naa50/San catalytic activity is essential for chromosome segregation during mitosis. Naa50/San N-terminally acetylates Scc1. NATs have distinct substrate specificities and their substrate recognition during protein translation depends on the nature of the first 2-5 amino acids of the elongating substrate polypeptide 7,10,31-33 . Analysis of the N-termini of all cohesin subunits showed that the N-terminal second and third amino acid residues of Scc1 were highly conserved across the eukaryotic tree of life (Fig. 5A). Since the Scc1 N-termini (MFY-) is compatible with the substrate specificity of Naa50/San 21,35 , we hypothesized that Naa50/San N-terminally acetylates Scc1. In contrast, the N-terminal sequences of the other cohesion subunits were less conserved and/or did not match the substrate specificity of Naa50/San (data not shown). In order to test this hypothesis we investigated if Drosophila Naa50/San was able to specifically N-terminally acetylate an Scc1 N-terminal peptide in vitro. We observed that the wild-type Naa50/San (San Wt ), but not the catalytically dead mutant (San R84A Y124F ), efficiently N-terminally acetylated a known substrate peptide of Naa50/San (positive control) and Scc1 N-terminal peptide (Fig. 5B). In contrast, Nt-Ac of a NatA substrate peptide and a Scc1 N-terminal peptide with a proline mutation that is known to block Nt-Ac 9 (negative controls) was poor (Fig. 5B). Multiple attempts to confirm that Scc1 Nt-Ac was also reduced in vivo after depletion of Naa50/San were inconclusive, as the N-terminal peptide of Scc1 was not detected by mass spectrometry (data not shown).

Naa50/San is not required for overall integrity of the cohesin complex.
Although the steady-state protein levels of Scc1 and Smc1 were normal in cells depleted for Naa50/San ( Supplementary Fig. 2C), Scc1 association to the mitotic metaphase chromosomes was nevertheless impaired (Supplementary Fig. 2A,B and D) 36,39 .
In order to investigate if depletion of Naa50/San impaired the overall integrity of the cohesin complex, endogenous Scc1 or a Myc-tagged Scc1 fusion protein expressed in Drosophila S2 cells were immunoprecipitated from actively dividing cells. All subunits of the cohesin complex, but significantly not Dalmatian/Sororin, were efficiently immunoprecipitated in both cases from total protein extracts prepared from control and from Naa50/ San-depleted cells ( Fig. 6A; detailed results are shown in Supplementary Table 4), suggesting that the overall integrity of the cohesin complex was not impaired after depletion of Naa50/San. Differences in the interaction between Scc1 and Smc3 in dividing or metaphase-arrested cells was similarly not detectably affected after depletion of Naa50/San from total protein extracts ( Fig. 6B,C). Since Dalmatian/Sororin was not immunoprecipitated with endogenous Scc1, Myc-tagged Scc1 (Fig. 6A) or GFP-tagged Smc3 (Fig. 6B,C), most of the isolated cohesin complex was most likely not interacting with sister chromatids in a fully cohesive-state. These results suggest that although depletion of Naa50/San impairs the correct establishment/ maintenance of cohesion, differences in interaction between Scc1 and Smc3 are not detectable by co-immunoprecipitation possibly because they were bridged together by Smc1.  for control RNAi and san RNAi-treated cells (96 hours after RNAi-treatment) was, respectively, 4.3% ± 0.2 (n = 1139) and 12.0% ± 0.7 (n = 983) (p < 0.001 Student's t-test). san RNAi but not control RNAi-treated cells showed single chromatids with their typical oscillatory behavior during metaphase (see arrowheads in B,D and E; see Movies S1-S6), which is suggestive of a loss of cohesion. (G) Frequency of cells with single chromatids after san RNAi-treatment was 5.6% ± 0.1 (n = 17) (72 hours after RNAi-treatment) (see Fig. 3) and 61% ± 11.7 (n = 77) (96 hours after RNAi-treatment). Single chromatids were never detected during mitosis and before anaphase in control RNAi-treated cells. Drosophila S2 cells stably expressed GFP-Histone H2B (green) and α-Tubulin-mCherry (red) (A,B) 70 or GFP-α -Tubulin (green) and CID-mCherry (red) (C-E) 71 . All images were obtained using maximum intensity projections of Z-stacks (0.8 μ m stacks of 5 sections each). For movies, Z-stacks were collected every 30 seconds. Scale bars equal 10 μ m. San protein levels after control RNAi and san RNAi-treatment are shown in Supplementary Fig. 3A.
The N-terminal second and third amino acid residues of Scc1 (MFY-) are highly conserved (Fig. 5A), and they are compatible with Naa50/San substrate specificity 21,35 . If Nt-Ac of Scc1 were one of the main reasons why Scc1 N-terminus is conserved, then a mutation capable of ensuring efficient Nt-Ac should only have a limited impact on Scc1 function. Consistently, ectopic expression of a mutant Scc1, where the second residue was changed to a Glutamate residue (from MF-to ME-; a NatB substrate likely to be fully Nt-Ac in vivo) ( Supplementary Fig. 3D), was still able to significantly suppress the mitotic arrest after depletion of Naa50/San (Fig. 7G). In contrast, ectopic expression of a mutant Scc1 protein whose N-terminal second residue was mutated to a Proline residue ( Supplementary Fig. 3D) (MF-to MP-; resulting in a proline N-terminus after methionine cleavage, which completely blocks protein Nt-Ac 9 ), failed to suppress the mitotic arrest and the single chromatids observed after depletion of Naa50/San (Fig. 7D and F; see quantification in Fig. 7G and H; see Movie S16; for Scc1 protein expression see Supplementary Fig. 3C). . In contrast, Naa50/San did not efficiently N-terminally acetylate a Naa10 substrate peptide SES-(negative control), nor the non-acetylatable N-terminal peptide MPY-. A catalytically dead-version of Naa50/ San (San R84A Y124F ) failed to efficiently N-terminally acetylate all tested peptides. Both wild type Naa50/San and Naa50/San R84A Y124F were heterologously expressed in E. coli (see methods). All measurements were done in the linear range of enzymatic activity (data not shown).
Scientific RepoRts | 6:39118 | DOI: 10.1038/srep39118 Ectopic expression of Scc1 (MP-) mutant protein induced, with or without depletion of Naa50/San, a major arrest in mitosis (Fig. 7G) and significantly increased the number of cells with loss of cohesion and single chromatids (Fig. 7A,C,D and F; see quantification in Fig. 7H; see Movies S15 and S16). The unexpected dominant negative effect of Scc1 (MP-) might result not only due to a complete blockage of Scc1 Nt-Ac but also from the conformational rigidity of proline and/or the expected loss of the Scc1 initiator methionine 66 . Either way, both possibilities highlight the functional importance of the first two N-terminal amino acids of Scc1 in the establishment of a stable interaction with Smc3 and for normal segregation of chromosomes during mitosis.

Discussion
We propose that in higher eukaryotes Naa50/San is required for mitotic sister chromatid cohesion by facilitating the correct interaction between Scc1 and Smc3. Yet, and in contrast with the nuclear Eco1/Deco lysine acetyltransferase whose role in Smc3 internal acetylation is important for the establishment of sister chromatid cohesion during DNA replication 36,49,50 , our results suggest that Naa50/San is most likely required but not instructive for the establishment/maintenance of cohesion. Although depletion of Naa60 (NatF) in Drosophila S2 cells results in the appearance of mitotic phenotypes 10 , adult Drosophila flies deleted for Naa60 are nevertheless viable (Rui Gonçalo Martinho, unpublished data) suggesting that Naa50/San is the dominant mitotic NAT. The in vitro substrate specificities of Naa50 and Naa60 are overlapping 10,33 , but these two enzymes appear to have non-overlapping in vivo substrates 67 .
Scientific RepoRts | 6:39118 | DOI: 10.1038/srep39118 potentially internally acetylate beta-Tubulin in vivo 41 , the kinetics of this reaction is most likely highly unfavourable as Naa50/San catalytic site cannot easily accommodate a side-chain lysine substrate 34 .
Naa50/San N-terminal acetylates Scc1 in vitro and most likely in vivo. Although the precise structural consequences of Scc1 Nt-Ac are still unknown and we failed to demonstrate that Scc1 is an in vivo substrate of Naa50/San, we propose that Naa50/San N-terminally acetylates the nascent Scc1 polypeptide during translation, which is potentially important for the correct folding of Scc1 N-terminal domain, and its subsequent interaction with Smc3 during the establishment of sister chromatid cohesion. Consistently, Scc1 N-terminus is highly conserved, its sequence is consistent with it being an in vivo substrate Naa50/San, and a mutation that blocks Nt-Ac of Scc1 impairs its function.
Why Naa50/San is not required for female germ line stem cell mitotic divisions is still unclear 40 , but it raises the possibility that the requirement of this enzyme varies during development, possibly due to redundancy with other NATs and/or tissue-specific regulators of sister chromatid cohesion. Alternatively, it is also possible that Nt-Ac of Scc1 is not always required for cohesion and Scc1 function. This is consistent with the observation that whereas Drosophila and human Naa50/San are required for sister chromatid cohesion, their orthologs in budding yeast are not 37,39 . Future work will clarify the molecular nature of the differential requirements of Naa50/San during Drosophila development.

Material and Methods
Fly work and genetics. Flies were raised using standard techniques. All Drosophila stocks used in this study are listed in Supplementary Table 1. The san alleles were isolated in a previously reported study 40 . Maternal mutant embryos and germ-line mutant clones were generated using the FLP/FRT ovoD system 69 . Germ-line clones of san 3 and san 4 were established by crossing FRT42B san/CyO virgins to hs-FLP; FRT42B ovo D /CyO males. The progeny was heat shocked twice at 37 °C for 90 min during second and third larval instar stages. As control we generated germ-line clones with FRT42B by crossing FRT42B/CyO virgins to hs-FLP; FRT42B ovo D / CyO males, followed by the heat shock procedure described before.
san loss-of-function mutant alleles 40 were complemented with a transgene carrying a genomic construct that contained a wild-type copy of the separation anxiety (san) gene locus with its own endogenous promoter (san wt ).
To check the requirement for Naa50/San catalytic activity, an identical genomic construct with two different amino acid mutations that rendered Naa50/San enzyme catalytically inactive (san R84A Y124F ) was also generated. Both constructs were similarly integrated in a genomic attP2 site. w; FRT42B, san 3 /CyO virgins were crossed to w; FRT42B, san 4 /CyO; san wt /TM6b males or to w; FRT42B, san 4 /CyO; san R84A,Y124F /TM6b males. Reciprocal crosses were also performed. Offspring were counted to determine viability of zygotic san mutant rescued by the two distinct genomic constructs. Maternal phenotypes were also analyzed. Embryos laid by w; FRT42B, san 3 / FRT42B san 4 ; san wt /+ females or by w; FRT42B, san 3 /FRT42B san 4 ; san R84A Y124F /+ were fixed and their early embryonic mitotic phenotypes analyzed. Complementation of san 3 germ-line clones (see above) 40,69 with san wt and san R84A Y124F genomic constructs was also performed. w; FRT42B san 3 /CyO; san wt /TM6B and w; FRT42B san 3 / CyO; san R84A Y124F /TM6B virgins were crossed with hs-FLP; FRT42B ovo D /CyO males, followed by the heat shock procedure as described above.
Genetic interaction studies: adult wings. The Gal4/UAS system 42 was used with the nubbin-Gal4 driver for tissue-specific expression in the blade regions of the larvae wing disc 43,44 . For adult wings genetic interactions studies five virgins w; nubbin-GAL4 UAS-san RNAi/CyO or five virgins w; nubbin-Gal4 were mated with 5-7 males from fly stocks containing distinct UAS-RNAi constructs or mutant alleles for genes of interest. The non-CyO progeny was scored into five distinct phenotypic classes accordingly to adult wings abnormalities (Fig. 1C). The crosses were repeated three times and the phenotypic class of more than thirty flies was evaluated for each cross.

Double-stranded RNA interference of Drosophila S2 cells. S2 cells were cultured at 25 °C and RNAi
was performed according to standard procedures. To deplete Naa50/San (encoded by san/CG12352) or Deco (encoded by deco/CG8598), S2 cells were transfected with double stranded RNAs (dsRNAs) corresponding to approximately 300-400 base pair fragments of each gene. To simultaneously deplete Naa50/San and Deco, S2 cells were simultaneously transfected with both dsRNAs. dsRNAs for GFP was used as control (sequence of used primers is shown in Supplementary Table 2). Each primer incorporates a T7 RNA polymerase-binding site. PCR products were used as template to synthesize dsRNA using the T7 RiboMAX Express kit (Promega). Drosophila S2 cells were counted and diluted to 2.5 × 10 6 cells/ml in serum free medium (SFM) supplemented with L-glutamine. Cells were incubated during 1 hour with 40 μ g for each dsRNA at a concentration of 10 μ g/ml. After 1 hour incubation with dsRNA, 3 ml of complete media was added back. S2 cells grew in the presence of the diluted dsRNAs and were analyzed 72 hours (Fig. 3) and 96 hours (Fig. 2) after dsRNA treatment. The double amount of control dsRNA was used in Fig. 3 to control for the total amount of dsRNAs used in the co-treatment with san and deco dsRNAs.
Live-cell imaging of Drosophila S2 cells. Live-cell imaging of S2 cells was done using S2 cells stably expressing GFP-Histone H2B and mCherry-α -tubulin 70 or GFP-α -tubulin and mCherry-centromere identifier (CID, a kinetochore marker) 71 (kindly provided by Helder Maiato (IBMC, Portugal)). The cells were cultured for 72 hours (in Fig. 3) or 96 hours (in Fig. 2) after RNAi-treatment, as described above. Cells were resuspended and plated in MatTek plates (P35G-1.5-20 C) pre-coated with Concanavalin-A 0.25 mg/ml (C2010; Sigma) 2 hours before observation. Visualization of live cells was performed using a Delta Vision Core System (Applied Precision) using a 100x UplanSApo objective and a cascade 1 K EMCCD camera (Photometrics). Images were acquired for a period of 2 hours at a frame capture rate of one every 30 seconds and a series of z-sections separated by 0.8-μ m intervals using softWoRx (Applied Precision, Inc.). Deconvolution was performed using the conservative ratio method in softWoRx software. Image sequences were converted to movies using the program ImageJ (http://rsb.info.nih. gov/ij/).

Transfection protocol of Drosophila S2 cells. S2 cells transfection with scc1 transgenic constructs was
done using FuGENE ® HD Transfection Reagent (E2311) (Promega). Cells were transfected using the reverse transfection protocol. A mix of 100 μ L of Serum Free Medium (SFM), 400 ng of DNA, and 4 μ L of Fugene HD was incubated at room temperature for 15 min. Meanwhile, cells were plated into 6 well plates at a concentration of 2.5 × 10 6 cells/well in serum free medium. The mix was added to the cells. After 4 h incubation at 25 °C, complete medium (with 10% Fetal bovine Serum) was added to stop transfection. Cells were incubated for 48 hours at 25 °C before starting the double-stranded RNA interference experiments (described above).

Generation of constructs and cloning.
Drosophila Scc1 open-reading frame (ORF) was obtained from a full-length Scc1 cDNA (clone FI11703). Wild type and mutant Scc1 ORFs where the second N-terminal amino acid residue was mutated to a Glutamate or Proline were cloned into pDONR221 (gateway system, Invitrogen). The N-terminal second residue point mutations were performed using the primers described in Supplementary Table 2. The wild type and mutant scc1 (known in Drosophila as verthandi) open reading frames were subcloned into a pHW vector with the Hsp70 promoter (gateway system, Invitrogen).

Immunofluorescence microscopy. Drosophila S2 cells.
After treatment with dsRNAs (see above), 2 × 10 6 cells were added to coverslips by 1 hour incubation at 25 °C. Cells were fixed with 4% formaldehyde, 0.03 M PIPES, 0.11 M HEPES, 0.01 M EGTA and 4 mM MgSO 4 for 10 min, followed by two washes in 1x PBS. Permeabilization and blocking was performed for 1 hour with PBS-TB (PBS, 0.1% Triton X-100, 1% Albumin from bovine serum). Primary antibody incubations were done in blocking solution for 2 hours at room temperature or overnight at 4 °C, followed by three 5 min washes in PBS-TB. Secondary antibody incubations were performed as described for the primary antibodies, including three 5 min washes. Primary antibodies included mouse anti-α -tubulin (DM1A) at 1:500 (Sigma), rabbit anti-pSer10-Histone H3 at 1:500 (Upstate Biotechnology), rabbit anti-Scc1 at 1:2000 72 (kindly provided by Claudio Sunkel (IBMC, Portugal)). Secondary antibodies used were anti-mouse Alexa Fluor 488 and anti-rabbit Alexa Fluor 555 at 1:1000 (Jackson ImmunoResearch). DNA was stained with DAPI at 1:1000 (stock concentration 1 mg/ml), with the addition of 5 μ g/ml RNAse A. Visualization of fixed cells was performed using a Delta Vision Core System (Applied Precision) using a 100x UplanSApo objective and a cascade 1 K EMCCD camera (Photometrics). Images were acquired as a series of z-sections separated by 0.2-μ m intervals. Deconvolution was performed using the conservative ratio method in softWoRx software.
Drosophila embryos. Drosophila embryos were fixed and stained as described before 73 . Briefly, control and san mutant embryos were collected (0-2 or 0-6 hours after egg-laying), dechorionated with 50% bleach for 5 min, washed with water, and fixed for 40 min with gently shacking in 4 mL heptane, 0.125 mL 37% formaldehyde and 0.875 mL phosphate buffered saline (PBS, pH 7.4). After removal of aqueous phase and addition of 4 mL methanol, the embryos were vigorously shaking during 1 min for removal of embryonic vitelline membrane. Following rehydration, embryos were blocked overnight at 4 °C with PBS containing 0.1% Tween-20, 0.1% bovine serum albumin and 1% donkey serum (BBT + serum). Primary antibody incubations were done overnight in BBT + serum at 4 °C. Embryos were washed extensively in PBS containing 0.1% Tween-20 (PBT), re-blocked in BBT + serum, and incubated with the appropriate secondary antibody for 2 hours at room temperature. Used primary antibody was anti-pSer10-Histone H3 at 1:1000 (Upstate Biotechnology). Secondary antibody was anti-rabbit Alexa Fluor 488 at 1:1000 (Jackson ImmunoResearch). Embryos were extensively washed in PBT and DNA was stained with DAPI at 1:10000 (stock concentration 2 mg/ml). Embryos were mounted in Fluorescent Mounting Medium (DakoCytomation) and were visualized using a LSM710 Confocal microscope. The Z-stacks projections were obtained using Image J program (Grouped Zprojector, maximum pixel intensity).

Biochemistry. Protein co-immunoprecipitation.
To analyze cohesin complex composition in S2 cells, co-immunoprecipitation was done using protein extracts from S2 cells expressing or not Myc-tagged Scc1. Briefly, for S2 cells with endogenous levels of Scc1, 1 mg of protein extract was diluted in 1 ml NB buffer and incubated with rabbit anti-Scc1 (1:250 dilution) or the pre-immune (1:10,000 dilution) as control, during 1 hr at 4 °C. Subsequently, 0.9 mg of Dynabeads Protein G (Invitrogen, Grand Island, NY, USA) were added to the immune complex and incubated 1 hr at 4 °C. For S2 cells ovexpressing Myc-tagged Scc1, 1 mg of protein was diluted in 1 ml NB buffer and incubated with 0.25 mg anti-c-Myc Magnetic beads (Invitrogen, Grand Island, NY, USA) for 1 hr at 4 °C. Both Dynabeads Protein G and anti-c-Myc Magnetic beads were then washed three times with NB buffer and resuspended in 50 μ l of ammonium bicarbonate (50 mM, pH 7.8). Samples were then analyzed by liquid chromatography coupled to tandem mass spectrometry (Mass Spectrometry Laboratory, Institute of Biochemistry and Biophysics, Poland).
For pull down assay from protein extracts of san-RNAi or control-RNAi treated S2 cells overexpressing GFP-tagged Smc3 (Smc3-GFP), and treated (Fig. 6B) or not (Fig. 6A) with 25 μ M of colchicine for 12 hours, 1 mg of protein was diluted in 1 ml NB buffer and incubated with 0.25 mg GFP-Trap Magnetic beads (Chromotek, Grand Island, NY, USA) for 1 hr at 4 °C. Beads were then washed three times with NB buffer and than boiled in 75 μ L of Laemmli buffer. The expression levels of Scc1, Dmt, San and Smc3 were than analysed by western blot analysis.
Mass spectrometry. Peptides mixtures were analyzed by LC-MS-MS/MS (liquid chromatography coupled to tandem mass spectrometry) using Nano-Acquity (Waters, Milford, MA, USA) LC system and Orbitrap Velos mass spectrometer (Thermo Electron Corp., San Jose, CA, USA). Prior to analysis, proteins were subjected to standard 'in-solution digestion' procedure, during which proteins were reduced with 100 mM DTT (for 30 min at 56 °C), alkylated with 0,5 M iodoacetamide (45 min in darkroom at room temperature), and digested overnight with trypsin (sequencing Grade Modified Trypsin-Promega V5111). The peptide mixture was applied to an RP-18 precolumn (nanoACQUITY Symmetry C18-Waters 186003514) using water containing 0,1% TFA as mobile phase, then transferred to nano-HPLC RP-18 column (nanoACQUITY BEH C18-Waters 186003545) using an acetonitrile gradient (0-35% AcN in 180 min) in the presence of 0.05% formic acid with a flow rate of 250 nl/min. The column outlet was directly coupled to the ion source of the spectrometer, operating in the regime of data dependent MS to MS/MS switch. A blank run ensuring no cross contamination from previous samples preceded each analysis.
Raw data were processed by Mascot Distiller followed by Mascot Search (Matrix Science, London, UK, on-site license) against Flybase database. Search parameters for precursor and product ions mass tolerance were 100 ppm and 0.6 Da, respectively, enzyme specificity: trypsin, missed cleavage sites allowed: 0, fixed modification of cysteine by carbamidomethylation, and variable modification of methionine oxidation. Peptides with Mascot Score exceeding the threshold value corresponding to < 5% False Positive Rate, calculated by Mascot procedure, and with the Mascot score above 30 were considered to be positively identified.
Human orthologs were determined using DSRC Integrative Ortholog Prediction Tool (DIOPT) (http://www. flyrnai.org/cgi-bin/DRSC_orthologs.pl). Only scores above two were considered such as the best matches when there was more than one match per input.
Western blot analysis. S2 cells were collected after centrifugation (1200 rpm, 10 min at 4 °C). 0-2 hours after egg-laying Drosophila embryos were collected and dechorionated with 50% commercial bleach solution. Both S2 cells and embryos samples were homogenized in NB buffer (150 mM NaCl, 50 mM Tris-HCl pH 7.5, 2 mM EDTA, 0.1% NP-40, 1 mM DTT, 10 mM NaF, and EDTA-free protease inhibitor cocktail (Roche)), and centrifuged at 20000 g for 3 min at 4 °C. Supernatant was recovered and centrifuged twice more. Bradford protein assay (BioRad) was used to calculate extract protein concentration. Protein samples were boiled for 5 min in Laemmli buffer (Sigma) and 15 μ g (per lane) was loaded onto a SDS 6% or 12% acrylamide gel. Proteins were subsequently transferred onto Hybond-ECL membranes (Amersham) and Western blotting was performed using standard procedures. Briefly, the Hybond-ECL membrane was blocked overnight in 5% non-fat milk/PBT (0.1% Tween-20, 1x PBS) at 4 °C. Primary antibodies were incubated overnight at 4 °C, with shaking. Following extensive washes with PBT, secondary antibodies were incubated for 4 hours at room temperature. After extensive washes, the proteins of interest were detected with an ECL Plus western blotting detection system (GE Healthcare). Primary antibodies used were rabbit anti-San at 1:1000 dilution, rabbit anti-Scc1 at 1:250 dilution, guinea pig anti-Smc1 at 1:500 dilution 74 , rat anti-Dmt at 1:100 dilution 75 , mouse anti-GFP (Roche) at 1:500 dilution, mouse anti-alpha-Tubulin (Dm1A) at 1:50,000 dilution (Sigma). Secondary detection was performed with anti-rabbit, anti-mouse, anti-guinea pig and anti-rat HRP-conjugated antibodies used at a final concentration of 1:5000. Uncropped images of all protein blots can be found in Supplementary Fig. 7.
Generated antibodies. Anti-Scc1 and Anti-San rabbit polyclonal antibodies were raised against His-tagged recombinant proteins corresponding to amino acids 80-184 of Naa50/San and 561-715 of Scc1 (Metabion international AG, Germany). Both antibodies were affinity purified. Anti-Scc1 and Anti-San antibodies were validated by western blot of protein extracts from S2 cells depleted for Scc1 ( Supplementary Fig. 5) or San ( Supplementary Fig. 3A), respectively.
Alignment of Scc1 N-terminal sequences. Protein sequence of Scc1 in H. sapiens was used to identify orthologs from fifteen species (7 holozoans, 2 fungi, 1 amoebozoa, 3 plants, 1 excavate and 1 chromalveolata) representative of the eukaryotic tree of life. Reciprocal bidirectional protein BLAST approach was used to retrieve the sequences from publicly available genome databases NCBI. Geneious R7 software was used with default parameters for alignment of N-terminal protein sequences.
In vitro acetylation of Scc1. Expression and purification of MBP-San WT and MBP-San R84A Y124F . The pETM41-san R84A Y124F mutant plasmid was generated using Stratagene multisite quikchange kit with pETM41-san WT plasmid as template and the primers listed in Table 2. One Shot ® BL21 Star TM Chemically Competent E. coli cells were transformed with pETM41-san WT or pETM41-san R84A Y124F encoding the MBP-San-6xHis and the MBP-San R84A Y124F -6xHis proteins, and a 200 mL culture was grown to an OD 600nm of 0.6 (at 37 °C), followed by transfer to 18 °C and addition of IPTG to a final concentration of 1 mM. Cells were incubated at 18 °C in a shaker at 250 rpm for 18 hours, and harvested the following day by centrifugation at 3000 × g and 4 °C for 15 minutes. Cell pellets were resuspended in 15 mL lysis buffer (50 mM Tris-HCl, 0.3 M NaCl, 2 mM DTT, 1 tablet/50 mL Complete EDTA free protease inhibitor cocktail (Roche), pH 7.4) and applied to the French press. After French press, the cell lysate was centrifuged for 15 minutes at 15,000 × g. The supernatant containing the soluble protein fraction was added to a 2 × 1 mL HisTrap column (Amersham), washed with IMAC wash buffer (50 mM Tris-HCl, 0.3 M NaCl, 2 mM DTT, 20 mM imidazole, pH 7.4), and eluted with IMAC elution buffer (same as wash buffer, but with 300 mM imidazole). Fractions containing recombinant protein were combined and subjected to gel filtration on a HiLoad 16/60 Superdex 75 column (Amersham), and eluted with gel filtration buffer (50 mM Tris-HCl, 0.3 M NaCl, 1 mM DTT, pH 7.4). Protein purity was confirmed by SDS-PAGE and Coomassie staining, and concentration was checked by absorbance measurements at 280 nm.
Scientific RepoRts | 6:39118 | DOI: 10.1038/srep39118 DTNB based Nt-acetylation assay. DTNB reacts with free thiol groups to give the product NTB 2− , the concentration of which can be measured spectrophotometrically. Using the method of Thompson et al. 76 , slightly modified 77 , we quantified the formation of NTB 2− after reaction of DTNB with an acetyltransferase assay sample. The time course acetyltransferase assay was performed by incubating purified MBP-San (300 nM) in acetylation buffer (50 mM Tris-HCl, 1 mM DTT, 10% glycerol, 0.2 mM EDTA, pH 8.5) with 150 μ M substrate peptide (Biogenes) and 150 μ M acetyl-CoA (Sigma Aldrich). Reactions were stopped with two times the volume of quenching buffer (3.2 M guanidinium-HCl, 100 mM sodium phosphate dibasic pH 6.8) after 20, 30, 40 and 60 minutes at 37 °C. For comparing San wt and San R84A Y124F activity, 300 nM enzyme was used in the same condition as above, only the reaction was stopped after 30 min. To measure CoA production, DTNB (2 mM final, dissolved in 100 mM sodium phosphate dibasic pH 6.8 and 10 mM EDTA) was added to the quenched reaction and the absorbance at 412 nm was measured. Thiophenolate production was quantified assuming λ = 13.7 × 103 M −1 cm −1 . Background absorbances were determined and subtracted from the absorbance determined for each individual reaction. Assays were performed in triplicate and turnover for the limiting substrate did not exceed 10%.